DESCRIPTION: This joint computational project between CARB and TIGR has three aims: 1) Initial investigation of the function of genes by means of sequence and database searches of evolutionary relationships. Established sequence search methods, as well as more advanced techniques, including Hidden Markov Models and threading, will be used. 2) Analysis of the experimentally derived structures to deduce function. The folds of new structures will be compared with known folds and their functions. The catalytic sties of enzymes as well as metal binding sites will be identified by comparison with a database of known three-dimensional motifs. Ligand binding sties will be identified with the aid of two different methods of surface topology analysis, electrostatic potentials, and by an examination of evolutionary conservation of the surface. Possible ligands will be identified by the use of docking techniques developed for drug design, together with a library of natural ligands. Sites of interaction with other macromolecules will be identified by analysis of surface composition and evolutionary conservation. 3) Distribution and collection of information about the function of each protein studied. Information about function derived from the program will be made available rapidly via a Web site, and research groups likely to be interested in particular functions will be contacted.